Analysis On Confluence Of Active Rules In Active Database

Traditional database systems have not satisfied current application. The research on active databases developed so rapidly these years, and has been applied to so many areas. Active database has become an advanced issue in the database area.A significant drawback in active database application lies in the development of correct rule application. It is very difficult in genera) to predict how a set of active database rules will behave.At the beginning of the paper, we introduce research background, development and current situation, characteristics and architecture of active database, compare a few typical active rule system and discuss knowledge model and execution model of active database. Then we analyze the behavior characteristics of active rule and on the basis of this we give an algorithm to check confluence of active rules.The work in this paper is as follows:To provide a efficient tool for the rule analysis, we must get the information of three properties of rule behaviors: they are termination, confluence and observable determinism. In this paper, we analyze these properties and discuss the primary methods to analyze these properties.For the sake of the analysis of confluence of active rules, we introduce the model based on execution graph. On the base of this model, we give the method based on execution graph that is used to analyze the confluence and the condition that guarantees that the set of rules is confluent.We give a algorithm that checks the confluence of active rules with meta-rules. l;irst, we introduce meta-rules that are used to specify active rule interactions into active rules. Then we analyze the static characteristics and on the basis of analysis give the algorithm that checks the confluence andanalyze its complexity.